Drowning remains a significant cause of accidental deaths, especially among children. Many children are non-swimmers and die as a result of falling into pools or off of boats; however, many children and adults who are swimmers die either from panic, exhaustion, cramps, seizures or a combination thereof. Children may drown despite being supervised while swimming. The parents or other adults supervising the child may have “just looked away for a second” only to find the child drowned on the bottom of the pool.
Several attempts have been made to address water safety with various degrees of success. For the non-swimmer, such as a toddler, the Safety Turtle™ device (Terrapin Communications Inc., Ottawa, Canada) is a bracelet, which when submerged triggers a poolside alarm to activate and to notify that a person has fallen into the water. Although the Safety Turtle™ device is excellent at detecting a person falling into the water, it may not be suitable for a child who is allowed to play in the water because the Safety Turtle™ device will generally activate in the course of normal play whenever the child's arm is submerged and produce false alarms.
Another approach taken to prevent drowning is to place an alarm on the pool itself. When a pool sensor detects entrance into the pool an alarm is activated. This alarm may be useful if the pool is empty, but is not suited for use with a child who is allowed to play in the pool. This device may not be easily transferred from one pool to another and may not be suitable for use in lakes or oceans.
Japanese Patent Publication No. 02241890 proposes a necklace, which when submerged would inflate and pull the drowning person to the surface by his/her neck. This may present a possibility of strangulation from the device itself. Because the device uses compressed air, it may only be used once. In addition, the amount of compressed air to float a person to the surface may entail a substantial amount of weight. In addition, the necklace could float to the surface and the user's head (which may be unconscious) might still be under water.
U.S. Patent Application Publication 2004/0095248 to Mandel proposes a device that is worn as a headband. When the device is submerged for a predetermined amount of time, it produces an ultrasonographic signal to be detected by sensors in the side of the pool to notify of a drowning person. This device is configured to transmit signals that propagate through water and is apparently dependent on a poolside receiver to detect ultrasonographic signals reliably. U.S. Pat. No. 4,714,914 to Boe proposes a wearable device, which when submerged will activate (or deactivate) a radio frequency alarm. Both devices may be limited by the power of the RF transmitter and the tremendous decrease in range and reliability that occurs when transmitters send a signal through a water/air interface. Such devices also may also be affixed to the body, for example, on a headband or on the back of the user. Both locations may be submerged for prolonged amounts of time when the user is not actually at risk for drowning. Therefore, false alarms remain a problem for these devices.
Other devices, such as U.S. Pat. No. 5,097,254 to Merrithew, depend on a pressure sensor to detect submersion for prolonged amounts of time. Pressure sensors may present a reliability problem because the difference in pressure difference between 3 inches below water and 18 inches below water are small and difficult to accurately detect or calibrate. However, even if calibrated correctly, a device that is six inches under water could indicate normal activity or it could indicate a drowning situation depending on where the device is worn, how long it has been submerged, etc. The calibration of such a device may become inaccurate over time due to normal wear on the device or changes in temperature.
Lifeguards, although not perfect, are a relatively reliable method of preventing drowning. Any device meant to augment drowning prevention must have a fail-safe design with a malfunction rate approaching zero. Previous attempts as described in the above art often rely on batteries, circuit boards, and sensors, all of which have a predefined failure rate which over time is unacceptably high. The algorithms described in our device have reduced those failure rates to a frequency approaching zero.
Accordingly, there remains a need for a reliable device for detecting potential drowning in users such as children who are permitted to have some water contact during the course of normal activities or play.